http://bugzilla.novell.com/show_bug.cgi?id=495417
User holler@nefkom.info added comment
http://bugzilla.novell.com/show_bug.cgi?id=495417#c5
Hans-Peter Holler changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|CLOSED |REOPENED
Version|Final |Factory
Component|Basesystem |Basesystem
CC| |holler@nefkom.info
AssignedTo|werner@novell.com |bnc-team-screening@forge.pr
| |ovo.novell.com
Resolution|WORKSFORME |
Product|openSUSE 11.0 |openSUSE 11.2
OS/Version|openSUSE 11.0 |SuSE Other
--- Comment #5 from Hans-Peter Holler 2009-04-17 14:07:13 MDT ---
Werner:
Sorry for reopening and hijacking, but maybe this is the fastest way :-)
rpm -q aaa_base-11.2-19.1 -changelog
* Do Apr 16 2009 werner@suse.de
- Make boot.clock more tough due udev timings (bnc#492921)
- Remove /etc/adjtime in boot.clock if left over (bnc#495417)
bnc#492921 isn't public, but maybe the explanation for the two hour time shift
on boot _before_ /etc/init.d/boot.clock is run after today's factory updates
can be found there.
What are these "udev timings", mentioned in the changelog?
My / my computer's problem:
On boot system _and_ hardware time is incremented +2 hours:
Extract from /var/log/boot.msg:
Kernel log daemon terminating.
Boot logging started on /dev/char/../tty1(/dev/console) at Fri Apr 17 18:01:09
2009
Trying manual resume from
/dev/disk/by-id/scsi-SATA_WDC_WD3200BEVT-_WD-WXE608HE8717-part3
Invoking userspace resume from
/dev/disk/by-id/scsi-SATA_WDC_WD3200BEVT-_WD-WXE608HE8717-part3
resume: libgcrypt version: 1.4.1
Trying manual resume from
/dev/disk/by-id/scsi-SATA_WDC_WD3200BEVT-_WD-WXE608HE8717-part3
Invoking in-kernel resume from
/dev/disk/by-id/scsi-SATA_WDC_WD3200BEVT-_WD-WXE608HE8717-part3
Waiting for device
/dev/disk/by-id/scsi-SATA_WDC_WD3200BEVT-_WD-WXE608HE8717-part4 to appear: ok
fsck 1.41.4 (27-Jan-2009)
[/sbin/fsck.ext3 (1) -- /] fsck.ext3 -a
/dev/disk/by-id/scsi-SATA_WDC_WD3200BEVT-_WD-WXE608HE8717-part4
/dev/disk/by-id/scsi-SATA_WDC_WD3200BEVT-_WD-WXE608HE8717-part4: clean,
372945/6103040 files, 13687185/24384661 blocks
fsck succeeded. Mounting root device read-write.
Mounting root /dev/disk/by-id/scsi-SATA_WDC_WD3200BEVT-_WD-WXE608HE8717-part4
mount -o rw,defaults -t ext3
/dev/disk/by-id/scsi-SATA_WDC_WD3200BEVT-_WD-WXE608HE8717-part4 /root
boot/91-killudev.sh: line 17: kill: (614) - No such process
Boot logging started on /dev/char/../tty1(/dev/console (deleted)) at Fri Apr 17
20:01:12 2009
done
got request to load boot.udev 4
got request to load startpar 5
startpar not found
Server closed connection
Starting udevd: done
Loading drivers, configuring devices: done
got request to load boot.loadmodules 4
boot.loadmodules not found
Loading required kernel modules
doneActivating swap-devices in /etc/fstab...
doneStart of /etc/init.d/boot.clock
System Time is Fri Apr 17 20:01:16 CEST 2009
System Time is Fri Apr 17 20:01:16 CEST 2009
The System Time is already in sync with Hardware Clockunused
Please have a look at the lines "Boot logging started on ...". You will see the
two hours difference.
The extra lines "Start of /etc/init.d/boot.clock
System Time is Fri Apr 17 20:01:16 CEST 2009
System Time is Fri Apr 17 20:01:16 CEST 2009"
come from /etc/init.d/boot.clock tweaked by me.
I have no clue what happens between these "Boot logging started on ..." lines.
As you can see these lines are written before /etc/init.d/boot.clock is run.
Wild guess for the two hours: UTC +1 for Germany/Berlin +1 for summertime.
No /etc/adjtime.
/etc/sysconfig/clock:
## Path: System/Environment/Clock
## Description: Information about your timezone and time
## Type: string(-u,--utc,--localtime)
## ServiceRestart: boot.clock
#
# Set to "-u" if your system clock is set to UTC, and to "--localtime"
# if your clock runs that way.
#
HWCLOCK="--localtime"
## Description: Write back system time to the hardware clock
## Type: yesno
## Default: yes
#
# Is set to "yes" write back the system time to the hardware
# clock at reboot or shutdown. Usefull if hardware clock is
# much more inaccurate than system clock. Set to "no" if
# system time does it wrong due e.g. missed timer interrupts.
# If set to "no" the hardware clock adjust feature is also
# skipped because it is rather useless without writing back
# the system time to the hardware clock.
#
SYSTOHC="no"
## Type: string(Europe/Berlin,Europe/London,Europe/Paris)
## ServiceRestart: boot.clock
#
# Timezone (e.g. CET)
# (this will set /usr/lib/zoneinfo/localtime)
#
TIMEZONE="Europe/Berlin"
DEFAULT_TIMEZONE="Europe/Berlin"
Hopefully you are able to reproduce this and know the one (you?) who could
resolve this.
Thanks in advance.
--
Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.